Whisper large-v3

OpenAI

Not hosted — mock backend Open weights

Sovereignty and hosting

Not hosted — mock backend

Not hosted on LLM EU on this deployment: requests are answered by the built-in mock backend in region dev-mock. No weights run.

Summary

Whisper large-v3 is OpenAI’s multilingual speech-recognition and speech-translation model with 1.54B parameters under the Apache-2.0 licence. It has a fixed 30-second receptive field and processes longer audio with a sliding window, and it uses 128 mel frequency bins rather than the 80 of earlier versions. The model card warns that it was evaluated on roughly ten languages and should be evaluated in-domain before deployment.

Specifications

Provider
OpenAI (US)
Context window
Not verified: the provider has not documented this claim.
Parameters
1.54B
Modalities
audio
License
apache-2.0 · licence
Open weights
yes
Release date
07 Nov 2023
Tasks
translation, multilingual-support, summarization

Languages

Strong: en, de, fr, es, it, pt, nl, pl

Adequate: sv, da, fi, cs, ro, el, uk, ru, tr, ar, zh, ja, ko, hi

Languages

Strong

  • en
  • de
  • fr
  • es
  • it
  • pt
  • nl
  • pl

Adequate

  • sv
  • da
  • fi
  • cs
  • ro
  • el
  • uk
  • ru
  • tr
  • ar
  • zh
  • ja
  • ko
  • hi

Not verified: the provider has not documented this claim. — a language is listed as strong only where a source claims it; the card links the evidence where one exists.

When to use it

Strengths

  • one model for transcription and translation to English
  • long-standing tooling support across every major inference stack
  • Apache-2.0 licence

Limits

  • fixed 30-second receptive field; longer audio needs chunking logic
  • the card only claims strong recognition for about ten languages
  • the card advises against using it for classification or speaker attributes